34 CORRELATION OF PHYSICAL FORCES. 



Another very usual mode of regarding the subject may 

 embarrass at first sight, but a little consideration will show 

 that it is explicable by the same doctrine. Water which has 

 ice floating in it will give, when measured by the thermometer, 

 the same temperature as the ice ; i.e. both the water and ice 

 contract the mercury of the thermometer to the point conven- 

 tionally marked as 32. It may be said, how is this recon- 

 cilable with the dynamical doctrine, for according to that 

 the solid should take from the mercury of the thermometer 

 more repulsive power than the liquid ; consequently, the ice 

 should contract the mercury more than the water ? 



My answer is, that in the proposition as thus stated, the 

 quantities of the water, ice, and mercury are not taken into 

 consideration, and hence a necessary dynamical element is 

 neglected : if the element of quantity be included, this objec- 

 tion will not apply. Let the thermometer, for instance, con- 

 tain 13^ oz. of mercury, and stand at 100 ; if placed in contact 

 with an unlimited quantity of ice at 32, the mercury will sink 

 to 32. If the same thermometer be immersed in an unlimited 

 quantity of water at 32, the mercury sinks also to 32; not 

 absolutely, perhaps, because, however great the quantity of 

 water or ice, it will be somewhat raised in temperature by the 

 warmer mercury. This elevation of temperature above 32 

 will be smaller in proportion as the quantity of water or ice 

 is larger than the quantity of mercury ; and, as we know of no 

 intermediate state between ice and water, the contact of a 

 thermometer at a temperature above the freezing point with 

 any quantity of ice exactly at the freezing point would, theo- 

 retically speaking, liquefy the whole, provided it had sufficient 

 time ; for as every portion of that ice would in time have its 

 temperature raised by the contact of the warmer body, and as 

 any elevation of temperature above the freezing point liquefies 

 ice, every portion should be liquefied. Practically speaking, 

 however, in both cases, that of the water and of the ice, 

 when the quantity is indefinitely great the thermometer falls 

 to 32. 



Now place the same thermometer at 100 successively in 

 one oz. of water at 32, and in one of ice at 32 ; we shall 

 find in the former case it will be lowered only to 54, .and in 



